Exchange rate effect on carbon credit price via energy markets

Hongik University and University of Illinois

IPAM UCLA, Commodity Markets and thier Financialization

From European Union Emissions Trading System (EU ETS) factsheet

• “Cap and Trade” First large greenhouse gas emission trading scheme. Still largest.
• Covers 11,000 factories, power stations, etc.
• 31 Countries. 28 EU member states plus Iceland, Norway, and Liechtenstein

EU ETS factsheet

• Emission allowances (emission credits) are allocated at the beginning of a trading period
• 1st period: Jan 2005 - Dec. 2007
• 2nd period: Jan 2008 - Dec. 2012
• 3rd period: Jan 2013 - Dec. 2020
  • One allowance gives the holder the right to emit one tonne \( CO_2 \)
  • Proposed cap for 2020 represents a 20% reduction in greenhouse gas emissions (ghg) from 2005

• In 1st and 2nd trading period initial allocations were more or less given away free by governments
• From 3rd period initial allocations are to be auctioned off
• After initial allocations are made firms can trade these credits
• -> Carbon market and Carbon prices

\[ P_{carbon} = f(P_{relative price}(Exchange Rate), D_{export}(Exchange Rate)|\Omega) \]

Taking the total derivative with respect to the exange rate we see a substitution effect and a demand effect.

\[ \begin{align*} \frac{dP_{carbon}}{dExchange Rate} = \frac{dP_{carbon}}{dP_{Relative Price}} \frac{dP_{Relative Price}}{dExchange Rate} \\ + \frac{dP_{carbon}}{dD_{export}} \frac{dD_{export}}{dExchangeRate} \end{align*} \]

Carbon prices respond to exchange rates because:

1. Substitution effect: exchange rates influence the relative price of clean and dirty energy (coal is denominated in USD and natural gas is denominated in EURO).
2. Demand effect: a depreciated EURO increases demand for exports.

Figure 2

Figure 3

Table 2: Pretests for model selection.

Augmented Dickey Fuller unit root test

CT: constant and trend, CNT: constant and no trend

Table 2: Johansen trace tests

a is the rank of cointegration tested

Cointegration among energy variables

• estimate first step
  \( e_t = P_t^{natural gas} - \beta_0 - \beta_1P_t^{coal} \)

• Then use these residuals as an explanitory variable in a SVAR

\[ B\Delta p_t = \Gamma_0 + \alpha e_t + \Gamma_1 \Delta p_{t-1} + \Gamma_2 \Delta p_{t-2} ... + \Gamma_J \Delta p_{t-J} + \epsilon_t \]

where

\[ \Delta p_t = \begin{bmatrix} \Delta Euro/USD \\ \Delta Euro/SFr \\ \Delta Coal \\ \Delta Natural Gas \\ \Delta Carbon \end{bmatrix} \]

A reduced form verson of the SVAR is defined by

\[ \begin{align*} \Delta p_t = B^{-1}\Gamma_0 + B^{-1}\alpha e_t + B^{-1}\Gamma_1 \Delta p_{t-1} + B^{-1}\Gamma_2 \Delta p_{t-2} + ... \\ + B^{-1}\Gamma_J \Delta p_{t-J} + B^{-1}\epsilon_t \end{align*} \]

Restricting parameters of the B matrix, allows the identification of the structural distubance term, \( \epsilon_t \), from the reduced form disturbance term, \( u_t = B^{-1}\epsilon_t \).

\[ u_t = \begin{bmatrix} * & 0 & 0 & 0 & 0 \\ * & * & 0 & 0 & 0 \\ * & * & * & 0 & 0 \\ * & * & * & * & 0 \\ * & * & * & * & *\end{bmatrix} \begin{bmatrix} \epsilon_t^{Euro/USD} \\ \epsilon_t^{Euro/SFr} \\ \epsilon_t^{Coal} \\ \epsilon_t^{Natural Gas} \\ \epsilon_t^{Carbon} \\ \end{bmatrix} \]

• We used a SVAR to show a link between carbon pricing and exchange rates

• A depreciated Euro caused lower carbon prices through energy substitution

  • \( \downarrow \) Euro \( \Rightarrow \) coal is expensive \( \Rightarrow \) substitute natural gas for coal
  • Firms regulated under the EU ETS are exposed to exchange rate risk and large players might need to consider hedging this risk in currency markets.